Professional Documents
Culture Documents
Workshop Presentations Fault-Induced Delayed Voltage Recovery (FIDVR)
Workshop Presentations Fault-Induced Delayed Voltage Recovery (FIDVR)
DOE-NERC Workshop on
Fault-Induced Delayed Voltage
Recovery (FIDVR) & Dynamic Load
Modeling
Overview
Page 1
10-7-10
Agenda
Wednesday, September 30, 2015
8:30-9:00 Welcome & Opening Remarks
David Meyer, U.S. Department of Energy, Office of Electricity Delivery and Energy
Reliability
David Till, North American Electric Reliability Corporation
9:00-9:15 Workshop Overview & Objectives
Joe Eto, Lawrence Berkeley National Laboratory
9:15-9:45 Current State of Load Modeling
A landscape overview of dynamic load modeling and FIDVR – where we are today, how we
got here, and where we’re going.
Dmitry Kosterev, Bonneville Power Administration
9:45-10:00 Break
10:00-12:00 Fundamentals, Testing & Modelings of Air-Conditioners
A deep dive into the fundamentals of motors, laboratory testing of end-use loads, and
modeling efforts. Development of single-phase and equivalent models using field testing
and detailed modeling.
John Undrill, Independent Consultant
Dmitry Kosterev, Bonneville Power Administration
Steven Robles, Southern California Edison
Bernie Leseiutre, University of Wisconsin
12:00-1:00 Lunch – provided
Agenda
Wednesday, September 30, 2015 (continued)
1:00-2:30 Manufacturing Perspective, Future Trends & Technologies (Panel Session)
Perspectives from the manufacturing community focusing on current and future trends in
control design and engineering, end-use requirements, and future technologies.
John Halliwell, Electric Power Research Institute
John Berdner, Enphase Energy
Tim Hawkins, Rheem
Hung Pham, Emerson Climate Technologies
2:30-2:45 Break
fidvr.lbl.gov
Joe Eto, LBNL
jheto@lbl.gov
Page 5
10-7-10
Composite Load Model
Development and Implementation
Presented by
Dmitry Kosterev, BPA
1
History Of Load Modeling
1980’s – Constant current real, constant impedance
reactive models connected to a transmission bus
Reflected the limitation of computing technologies of that time
2
1996 Large-Scale Outages in the West
Malin 500-kV Bus Voltage, July 2 1996 Outage
600
500
400
July 2, 1996
300
200
100
0
0 5 10 15 20 25 30 35 40
August 4, 2000
3
2001 “Interim” Load Model
2001 – WECC “Interim” Load Model:
• 20% of load is represented with induction motors, the
remaining load is static, mainly constant current active,
constant impedance reactive components
• Motors were connected at high voltage bus, data
representative of large fan motors (source John Undrill)
• Same percentage was applied to all areas in WECC
• Was the only practical option available in 2001
• “Interim” load model was intended as a temporary solution to
address oscillation issues observed at California – Oregon
Intertie
• Was in use until 2014 when superseded with composite load
model
4
Events of Delayed Voltage Recovery in Southern
California
• 1980’s – Southern California Edison observed events of
delayed voltage recovery attributed to stalling of
residential air-conditioners
– Tested residential air-conditioners, developed empirical AC
models
event. 520
500
480
• 2004-06 FIDVR events in
Valley area
460
440
420
400
-10 0 10 20 30 40
5
Southern California Edison
Need to represent a
distribution equivalent
B. Data
Tools for data management are available
Processes for providing data are established
Default data sets are available
C. Studies
Model validation studies
System impact and sensitivity studies 9
Composite Load Model Structure
Composite Load Model Structure
M
12.5-kV
13.8-kV M
69-kV
115-kV
M
138-kV
AC
UVLS
UFLS Electronic
GE PSLF
Siemens PTI PSS®E
Power World Static
11
“Performance Model” for Air-Conditioners
Real Power Reactive Power
6 6
5 5
3 3
2 2
RUN
0 0
0 0.2 0.4 0.6 0.8 1 1.2 0 0.2 0.4 0.6 0.8 1 1.2
Voltage (per unit) Voltage (per unit)
Motors stall when voltage drops below Vstall for duration Tstall
A fraction Frst of the aggregated motor can restart when the voltage exceeds Vrst for
duration Trst
12
Motor thermal protection is modeled
Data
13
Load Model Data
Load Component
Distribution Equivalent Data Model
Data
M
M
69-kV
115-kV
138-kV M
Load Model
Composition
M Data
Static
14
Load Model Data
• Develop understanding of electrical end uses in
various buildings, do not rely solely on
consultants with elaborate building models
– Building models can help develop understanding, but
should not be used as the primary source
• When you walk in Whole Foods on hot summer
day, do you know how much load is refrigeration
/ AC / lighting / fans / cooking? Do you know
expected size and type of compressor motors? Do
you know what building EMS system will possibly
do during a fault?
• We need to develop this expertise
15
Summer peak demand in California
Peak Demand (GW) Consumption (TWh)
8 35
7
30
6
Peak Demand
25
Annual Consumption
5
15
3
10
2
5
1
0 0
Res. - Air Com'l. - Air Com'l. - Com'l. - Other Res. - Res. - Com'l. - Res. - Res. - Dryer Com'l. -
Conditioning Conditioning Interior M iscellaneous Refrigerator Ventilation Cooking Refrigeration
Lighting
Residential AC
Lighting
Commercial AC Refrigeration
Ventilation
LBNL
Heat Exchanger Fans
Hotel in Salt Lake City
125 rooms
Compressor
2 Compressor Motors:
A: 3-ph, 460 V, 139 RLA, ~94kW / 70 hp
B: 3-ph, 460 V, 118 RLA, ~80kW / 60 hp
9 Fan Motors:
3-ph, 460V, 1.25 hp each
Compressor
CEC California Commercial End-Use Survey
Summer
< 1%4%
<<1%Peak Load
1%
5%
6%
33%
Cooling
Lighting
26%
Ventilation
11%
< 1%
9% 4%< 1%
Refrigeration
Residential Commercial
Temperature
110
100
90
80
70
60
50
0 3 6 9 12 15 18 21 24
30
25
20
15
10
0
0 3 6 9 12 15 18 21 24
12
10
0
0 3 6 9 12 15 18 21 19 24
Climate Zones
NWC – Northwest coast
NWC`
NWV – Northwest valley
RMN
NWI NWI – Northwest inland
NWV RMN – Rocky mountain
NCC – N. Calif. coast
NCV – N. Calif. Valley
NCC
HID
NCI – N. Calif. Inland
NCV HID – High desert
SCC – S. Calif. coast
SCV – S. Calif. Valley
SCV
SCC
SCI – S. Calif. Inland
DSW
DSW – Desert southwest
20
Substation / Feeder Types
• Residential: typical of your suburban neighborhood
• Commercial: typical of downtown load
• Mixed (default): mix of residential and commercial
• Rural / agricultural
• Several types of industrial loads (petro-chemical,
paper mill, steel mill, semiconductor, etc)
21
Load Composition Model
BPA and WECC
Load Profile
30
developed Load 25
Static P Current
Composition Model:
20
Static P Resistive
Power Electronic
15
12 climate zones X
Motor D
Motor C
10 Motor B
4 feeder types + 5
Motor A
5 seasons 0.25
24 hours
Motor A
0.20 Motor B
Motor C
Motor D
0.15
Power Electronic
Static P Resistive
0.10 Static P Current
0.05
22
0.00
0 4 8 12 16 20 24
Load Model Data Tool
Inputs:
- File with load records, including their “load type
identifier”
- Load composition data
- Motor and end-use model data
Output:
- PSS®E DYR and PSLF DYD model data records
24
Studies
Model Acceptance and Validation
Studies
Tens of thousands runs have been done with composite
load model up to date
Challenges:
- Load composition varies daily and seasonally
- Lack of disturbance recordings, particularly FIDVR
records outside Valley area in Southern California
26
August 4, 2000 Oscillation
545
WECC Interim Model 545
CMPLDW with default data
543 543
541 541
539 539
537 537
Voltage (kV)
Voltage (kV)
535 535
533 533
531 531
529 529
527 527
525 525
0 5 10 15 546 20 25 30 35 40 0 5 10 15 20 25 30 35 40
540
538
536
534
532
530
528
526
524
0 5 10 15 20 25 30 35 40
1.1
0.9 Actual
Simulated
0.8
0.7
0.6
5 10 15 20 25 30 35
Frequency
60.2
60.1
60
59.9
59.8 Actual
Simulated
59.7
5 10 15 20 25 30 35
Time (sec)
29
System Impact Studies
• We were able to tune the composite load model to
reproduce historic system events
That said…
• Composite load model was more conservative in
simulating the severity of FIDVR events than we
expected
• Another concern is high sensitivity of results with
respect to stall and motor protection assumptions
Therefore,
• WECC adopted phased implementation of composite
load model: Phase I – air-conditioner stalling feature
is disabled
30
Implementation
Phased Implementation in the West
2011 WECC approved phased implementation plan:
• Model indicated AC stalling much greater than what experienced
in reality outside Valley area in Southern California
• Lack of validation outside Valley area
• WECC voltage dip criteria
• More gradual transition
33
Where we are now …
• Composite load model is implemented in GE PSLF and
Siemens PTI PSS®E, similar models exist in Power World,
Power Tech TSAT
• Tools are developed for load model data management
• “Default” data sets are prepared
• WECC is taking phased approach for approving the
composite load model for TPL compliance studies
– Phase 1: air-conditioner stalling is disabled by setting Tstall
parameter to a large number
– Phase 2: better understand the reliability implications of
delayed voltage recovery due to air-conditioner stalling, develop
appropriate reliability metrics
… Where we are now
• All planning and seasonal operational cases prepared
by WECC now have composite load model
• Tens of thousands runs have been done with the
composite load model up to date
• WECC studies help to improve model data sets
Next Steps
AC Model Revisions
New work by John Undrill, Bernie Lesieutre and BPA suggests that
air-conditioners may not stall as easy as previously expected
37
Planned Revisions to Model Structure
Flexible model structure
Electrical end-use characteristics are changing rapidly, as more
loads become electronically connected
Modular structure (similarly to generating units)
Air-conditioner models
Revise “performance” model to reflect recent test findings
Add MOTORC dynamic model
Distributed generation
38
Model Data
Composite load model is a very powerful.
Building surveys
- Installed equipment
- Load shapes
- Protection and control
End-use monitoring
40
Load Modeling – Setting Expectations
We can now achieve the great accuracy with generator
models:
We model physical equipment that is well defined and under
our control
41
Thank You
42
End-Use Testing at BPA
Presented by
Dmitry Kosterev and Steve Yang, BPA
1
BPA End-Use Testing Lab
3.4 0.64
3.2 0.62
3 0.6
2.8 0.58
2.6 0.56
80 85 90 95 100 105 110 115 4
Ambient Temperature (F)
(a) Torque-Speed Curves
Auxiliary R
Winding Main
Winding
S
Run Thermal
Capacitor Relay
C
75 mm
310 mm
6
(c) Compressor Load Torque in very cyclical
Torque
360 720
Rotor Position
It is very possible that the motor stalls at the next compression cycle7
Compressor Motor Tests –
Power-Voltage Trajectories
Real Power
12000
115F
110F
10000 105F
STALL
100F
95F
Compressor Real Power [W]
8000 90F
85F
80F
6000
Reactive Power
RUN 12000
4000 115F
STALL 110F
10000 105F STALL
STALL
4000
* note motor load and
stall voltage increase 2000 RUN
with temperature
0
0 50 100 150 200 8
Voltage [V]
“Performance” Model
Real Power Reactive Power
6 6
5 5
3 3
2 2
RUN
0 0
0 0.2 0.4 0.6 0.8 1 1.2 0 0.2 0.4 0.6 0.8 1 1.2
Voltage (per unit) Voltage (per unit)
Motors stall when voltage drops below Vstall for duration Tstall
10
Testing and Modeling – Round 2
BPA, John Undrill and Bernie Lesieutre found a
“common mode failure” mode in SCE-BPA-EPRI testing -
all voltage sags were sudden and applied at voltage
waveform zero-crossing
11
AC Stall, Instantaneous Voltage Dip
Previously thought
Zero Crossing
Peak
45°
12
AC Stall, : 1 Cycle Ramp in Voltage Dip
Previously thought
Peak
13
Next Steps
1. Update AC “performance” model
15
1
NERC FIDVR and Dynamic Load Modeling Workshop, September 30, 2015.
Support from CERTS and PSERC
2
.
x
u -5
a
-10
“Performance” Model
-15
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
time
4
Tmech
Tave
Point-on-Wave Effects
Simulations of Single-Phase Compressor Motor
Applied voltage. The disturbance occurs
at different points along the sinusoid:
peak, zero crossing, in between.
Point-on-Wave Effects
Stall Voltage vs fault duration,
and point-on-wave variation.
Point-on-Wave Effects
Point-on-Wave Effects
• These results suggest a reason why FIDVR
events don’t cascade beyond an event feeder.
• Locally, A/C motors stall in response to event.
Further away, the filtered
voltage may exceed threshold.
Performance Model Characteristic for
reference
Laboratory Tests
• Air Conditioner Tests at BPA Facility
• Test Point-on-Wave Response, with and
without ramp.
• Scroll Compressor
Voltage dip to 48, 45, 40, 35 and 30% nominal
Recovery voltage at 90% nominal.
Zero Crossing
Peak
45°
12
Peak
Do All Motors Stall?
Grid Voltage
V
Grid Voltage
V
Conclusions
CONFIDENTIAL
2 Southern California Edison
Laboratory Setup
• Grid Simulator • Digital Oscilloscope
‒ Voltage Probes
• Equipment under test ‒ Current Transformers (CTs)
‒ 3Ф Commercial Rooftop A/C Unit
‒ Thermocouples
‒ Residential A/C Unit with VFD
‒ Accelerometers
CONFIDENTIAL
3 Southern California Edison
Tests Performed
• Compressor Shutdown
• Compressor Startup
• Balanced Under/Over-Voltage Transients
• Unbalanced Under/Over-Voltage Transients
• Under/Over-Frequency Transients
• Voltage/Frequency Oscillations
• Voltage/Frequency Ramps
• Harmonics
• Conservation Voltage Reduction
CONFIDENTIAL
4 Southern California Edison
Commercial 3Ф Rooftop A/C Testing
CONFIDENTIAL
5 Southern California Edison
3Ф A/C Contactor Dropout Summary
• Dropout is dependent on voltage supplying the A/C
unit controls (VФA-ФB, VФB-ФC, or VФC-ФA)
CONFIDENTIAL
6 Southern California Edison
3Ф A/C Stalling Results Summary
• Contactor drops out before stalling can occur for
3-phase balanced under-voltage conditions
CONFIDENTIAL
7 Southern California Edison
3Ф A/C Stalling Results (Sample)
• Compressor performance during unbalanced under-
voltage (Phases A & B) transients:
‒ Stalled at 20% VL-N in 24 cycles
‒ Stalled at 10% VL-N in 12.6 cycles
CONFIDENTIAL
8 Southern California Edison
3Ф A/C Stalling (Modified Units)
• Units were modified such that controls were powered separately
to bypass dropout
• Captured I, P, and Q at different balanced voltage levels, including
at the stalling and restarting point
CONFIDENTIAL
9 Southern California Edison
Residential VFD A/C Testing
CONFIDENTIAL
10 Southern California Edison
VFD A/C Compressor Startup
• VFD A/C units display low inrush current compared to
conventional units
‒ Largest inrush current: 11.3 Amps within 1.8 cycles
‒ Compressor current ramps up in 20 – 50 seconds
‒ Consumption increases periodically to meet temperature demand
VFD A/C #1
11.3 A
CONFIDENTIAL
11 Southern California Edison
VFD A/C Controls Dropout Summary
• No stalling behavior was observed
CONFIDENTIAL
12 Southern California Edison
VFD A/C Controls Dropout Summary
130 cycle Transients 12 cycle Transients 3 cycle Transients
Unit t trip / dropout t trip / dropout t trip / dropout
V trip / dropout (%) V trip / dropout (%) V trip / dropout (%)
(cyc) (cyc) (cyc)
VFD A/C #1 52% 15 55% 13.2 51% 3.6
VFD A/C #2 56% 7.8 55% 9 N/A N/A
VFD A/C #3 58% 130.8 58% 12.6 59% 4.2
VFD A/C #4 69% 130.1 49% 12.9 N/A N/A
VFD A/C #5 41% 130.8 20% 12.6 N/A N/A
VFD A/C #6 81% 131.1 56% 12.2 N/A N/A
VFD A/C #7 62% 129.6 60% 12 54% 4.2
CONFIDENTIAL
13 Southern California Edison
VFD A/C Harmonics Contribution
• VFD A/C #1, 4, 5, 7 current THD is 11% - 16.9% of fund.
• VFD A/C #3 current THD is ~29% of fund.
• VFD A/C #2 & #6 current THD is 39% - 47.5% of fund
VFD A/C #6
CONFIDENTIAL
14 Southern California Edison
Thank You.
CONFIDENTIAL
15 Southern California Edison
FIDVR
The transient behavior of loads
John Undrill
September 2015
Load modeling issues
Load composition
lighting
electronic power supplies
single phase motors
three phase motors
Behavior in transients
stay-on/shutdown discharge and LED lighting
slow-down/reaccelerate miscellaneous motors
run/stall residential air conditioners
Asymptotic behavior
linear dynamics control gain and bandwidth
mode changes sensitivity to voltage/frequency conditions
Load modeling issues
ibs
0 0
pe
0 0
-5
Speed dips during fault -
-10 -5
2.9 3 3.1 3.2 3.3 3.4 2.9 3 3.1 3.2 3.3 3.4 reacceleration is decisive
10 1
Immediate negative peak of
8 0.8
6 0.6
torque transient approaches
speed
4 0.4
2 0.2
0 0
2.9 3 3.1 3.2 3.3 3.4 2.9 3 3.1 3.2 3.3 3.4
5 1
Well understood behavior
0.5
Central to circuit breaker
vas
0 0
te
rating standards
-0.5
-5 -1
2.9 3 3.1 3.2 3.3 3.4 2.9 3 3.1 3.2 3.3 3.4
Voltage dip at terminals 10 10
ias
ibs
0 0
-5 -5
10 5
ics
pe
0 0
components -5
-10 -5
2.9 3 3.1 3.2 3.3 3.4 2.9 3 3.1 3.2 3.3 3.4
Reactive power reverses during voltage
dip - motor contributes to support of 5 1
0.8
voltage
0.6
speed
qe
0
0.4
Immediate negative peak of torque 0.2
vas
0 0
te
-5 -1
2.9 3 3.1 3.2 3.3 3.4 2.9 3 3.1 3.2 3.3 3.4
Air conditioner rotor - approximately 5kW
ias
ibs
motor driving a 0 0
-200 -200
1.9 2 2.1 2.2 2.3 2.4 1.9 2 2.1 2.2 2.3 2.4
H = 0.048 second
1 25
0.8 20
Speed is pulled down very
speed
0.6 15
strongly by the negative
tl
0.4 10
electromagnetic torque
0.2 5
0 0
Motor stalls and does not 1.9 2 2.1 2.2 2.3 2.4 1.9 2 2.1 2.2 2.3 2.4
restart
50 500
of torque transient
vas
-50 0
te
0 0
vt
it
-100
-200 -500
1.9 2 2.1 2.2 2.3 2.4 1.9 2 2.1 2.2 2.3 2.4
FIDVR is not only a positive sequence issue
320
300
280
260
240
Vl-n/V
220
200
180
160
140
120
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
Motor behavior is sensitive to
100 1.2
0.8
0
Speed, pu
Telec nm
0.6
-50
characteristics -100
0.4
0.2
(torque/speed/angle) -150 0
-200 -0.2
1.95 2 2.05 2.1 2.15 2.2 2.25 1.95 2 2.05 2.1 2.15 2.2 2.25
Terminal votage, V
0.4
Tload, pu 15 0.2
initiation of voltage dip 0
10 -0.2
-0.4
• Inverter capabilities
• Old Inverters
• Smart Inverters
• Smart inverter functions defined today
• Defining the needed FIDVR response
• Normal operation (steady state) versus FIDVR (transient)
• Conclusions
2
2| © 2015 Enphase Energy, Inc. | CONFIDENTIAL
The Old Days (2000 - 2003)
3
3| © 2015 Enphase Energy, Inc. | CONFIDENTIAL
Today (2014 - 2016)
• CA rule 21 approves smart inverter functionality.
Phase 1 autonomous behaviors (Dec 2015)
• Voltage and frequency ride through
• Real and reactive power control
• Return to service behaviors / ramp rate control
• Hawaiian Electric Inc. implements mandatory ride
through requirements (Jan 2015)
• CA rule 21 Phase 2 in development.
• IEC 61850 data model, IEEE 2030.5 / SEP 2.0 Protocol
• Updates to interconnection handbooks under development
• Relevant Standards
• UL 1741, UL 1741 Supplement A, IEEE 1547, 1547.a,1547.1
• IEC 61850, IEEE 2030.5
• UL 1998 (firmware certification)
4
4| © 2015 Enphase Energy, Inc. | CONFIDENTIAL
New Regulatory Concepts (in the US)
5
5| © 2015 Enphase Energy, Inc. | CONFIDENTIAL
UL 1741 Supplement A Functions
8
8 | © 2015 Enphase Energy, Inc. | CONFIDENTIAL
PV Inverter Operating Areas
9
9 | © 2015 Enphase Energy, Inc. | CONFIDENTIAL
Terminology of FIDVR Response
• “Event based dynamic reactive current support” (EPRI)
• Provide capacitive reactive current in response to low voltage
• Similar to EPRI VV12 but transient in nature
• Reduce active power to supply reactive power (VAR Priority)
Courtesy of EPRI, Common functions of smart inverters, Version 3, Feb, 2014, Brain Seal
10
10 | © 2015 Enphase Energy, Inc. | CONFIDENTIAL
Dynamic Reactive Current Support (EPRI)
DbVMin This is a voltage deviation relative to Vaverage, expressed in terms of % of Vref (for example -10%Vref).
For negative voltage deviations (voltage below the moving average) that are smaller in amplitude than this
amount, no additional dynamic reactive current is produced.
DbVMax This is a voltage deviation relative to Vaverage, expressed in terms of % of Vref (for example +10%Vref).
For positive voltage deviations (voltage above the moving average) that are smaller in amplitude than this
amount, no additional dynamic reactive current is produced. Together, DbVMin and DbVMax allow for the
creation of a dead-band, inside of which the system does not generate additional reactive current support.
ArGraSag This is a gradient, expressed in unit-less terms of %/%, to establish the ratio by which Capacitive % VAR
production is increased as %Delta-Voltage decreases below DbVMin. Note that the % Delta-Voltage may be
calculated relative to Moving Average of Voltage + DbVMin (as shown in Figure 16-1) or relative to Moving
Average of Voltage (as shown in Figure 16-4), according to the ArGraMod setting.
ArGraSwell This is a gradient, expressed in unit-less terms of %/%, to establish the ratio by which Inductive % Var
production is increased as %Delta-Voltage increases above DbVMax. Note that the % Delta-Voltage may
be calculated relative to Moving Average of Voltage
+DbVMax (as shown in Figure 16-1) or relative to Moving Average of Voltage (as shown in Figure 16-4),
according to the ArGraMod setting.
FilterTms This is the time, expressed in seconds, over which the moving linear average of voltage is calculated to
determine the Delta-Voltage.
Courtesy of EPRI, Common functions of smart inverters, Version 3, Feb, 2014, Brain Seal
12
12 | © 2015 Enphase Energy, Inc. | CONFIDENTIAL
Optional Variables of DRCS
Additional Settings (Optional)
ArGraMod This is a select setting that identifies whether the dynamic reactive current support acts as shown in Figure
16-1 or Figure 16-4. (0 = Undefined, 1 = Basic Behavior (Figure 16-1), 2 = Alternative Behavior (Figure 16-
4).
BlkZnV This setting is a voltage limit, expressed in terms of % of Vref, used to define a lower voltage boundary,
below which dynamic reactive current support is not active.
HysBlkZnV This setting defines a hysteresis added to BlkZnV in order to create a hysteresis range, as shown in
Figure 16-5, and is expressed in terms of % of VRef.
BlkZnTmms This setting defines a time (in milliseconds), before which reactive current support remains active
regardless of how deep the voltage sag. As shown in Figure 16-5.
Enable/Disable Event-Based This is a Boolean that selects whether or not the event-based behavior is enabled.
Behavior
Dynamic Reactive Current Mode This is a Boolean that selects whether or not Watts should be curtailed in order to produce the reactive
current required by this function.
HoldTmms This setting defines a time (in milliseconds) that the delta-voltage must return into or across the dead-band
(defined by DbVMin and DbVMax) before the dynamic reactive current support ends, frozen parameters
are unfrozen, and a new event can begin.
Courtesy of EPRI, Common functions of smart inverters, Version 3, Feb, 2014, Brain Seal
13
13 | © 2015 Enphase Energy, Inc. | CONFIDENTIAL
Priority of Smart Inverter Functions
• Multiple functions can be running simultaneously
• Can lead to conflicting requirements
• Example: active power needed during under frequency versus
reactive power needed for voltage regulation / FIDVR
• What is priority of functions during FIDVR event ?
• 1) Frequency support of bulk system
• May cause limitations of reactive power capabilities (W priority)
• 2) FIDVR response ? (New concept needs discussion)
• 3) Steady state voltage regulation (FPF, V/VAr)
• 4) Commanded active / reactive power
• 5) Scheduled responses
14
14 | © 2015 Enphase Energy, Inc. | CONFIDENTIAL
Conclusions
• Smart Inverters can provide dynamic reactive power
in response to FIDVR events
• Capability exists today but functional requirements are TBD
• Regulatory standards are under development now and
FIDVR response is “on the agenda”
• IEEE 1547 (2016), IEEE 1547.1 (2016/170
• UL 1741 Supplement A (2015), Full revision (2016)
• Definition of the desired functionality is needed in
order to implement and certify
• Inverters are very flexible and behaviors can be complex
• Inverter models are very complex but will be critical in
determining best guesses for initial functionality
• Remote upgradability of inverters will likely be needed
as PV proliferates and understanding evolves
15
15 | © 2015 Enphase Energy, Inc. | CONFIDENTIAL
Thank you for your attention!
Smart PV Inverter
Smart System
Smart EV Charger (composite)
Smart Loads
1
17 | © 2015 Enphase Energy, Inc. | CONFIDENTIAL
7
Plug-in Electric
Vehicle Charging
Characteristics
John Halliwell
Principal Project Manager – Electric
Transportation
NERC-DOE FIDVR Workshop
September 30, 2015
Washington, D.C.
2
© 2015 Electric Power Research Institute, Inc. All rights reserved.
Some Electric Vehicle Supply Equipment Samples
Level 2 AC
208/240V charging
DC Fast
208/480V 3phase input
3
© 2015 Electric Power Research Institute, Inc. All rights reserved.
Plug-in Vehicle On-board System Topology
Battery
Management
PEV System DC
Traction Charger AC
Battery AC/DC
J1772™
Charge
Port
4
© 2015 Electric Power Research Institute, Inc. All rights reserved.
Topology and Nomenclature – AC Charging
PEV
Battery
Management
System
1-ph AC
Traction Charger AC EVSE Service
Battery AC/DC
120/208/240V
J1772
Charge
Port
5
© 2015 Electric Power Research Institute, Inc. All rights reserved.
DC Charging
3-ph AC
PEV Service
DC
EVSE 208/480V
DC
BMS
Traction Charger
Battery AC/DC
DC Charge Port
6
© 2015 Electric Power Research Institute, Inc. All rights reserved.
Wireless Charging
85kHz 1-ph AC
AC Wireless Service
EVSE 208/240V
PEV
DC
BMS
Traction
Charger
Battery
AC/DC
7
© 2015 Electric Power Research Institute, Inc. All rights reserved.
No Matter the Charging Technology…
8
© 2015 Electric Power Research Institute, Inc. All rights reserved.
Power versus Time
DC “FAST”
Up to 100+ kW
CHARGING
AC Level 2
Up to 19.2 kW
CHARGING
Wireless CHARGING
AC Level 1
CHARGING Up to 2 kW
9
© 2015 Electric Power Research Institute, Inc. All rights reserved.
EPRI 2011 Nissan Leaf Data
Roughly Constant
Power down to
208V
10
© 2015 Electric Power Research Institute, Inc. All rights reserved.
EPRI 2011 Chevy Volt
Roughly constant
power down to
208V
11
© 2015 Electric Power Research Institute, Inc. All rights reserved.
This Sample Waveform was Used
12
© 2015 Electric Power Research Institute, Inc. All rights reserved.
2013 Nissan Leaf – Response to FIDVR Voltage Profile
Voltage (V)
46.3s
Current (A)
Power (W)
6730W
6700W 6745W
13
6727W © 2015 Electric Power Research Institute, Inc. All rights reserved.
2013 Nissan Leaf – Response to FIDVR Voltage Profile - 2
Voltage
Reduction
Voltage (V)
Current (A)
14
© 2015 Electric Power Research Institute, Inc. All rights reserved.
Together…Shaping the Future of Electricity
15
© 2015 Electric Power Research Institute, Inc. All rights reserved.
Load Model Improvements
– a Case Study
Donald Glen Davies
Chief Senior Engineer
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
2
Data Management
• Complexity of load modeling
– Each load has a different composition
– Load magnitude and composition changes over a
day, over a year, and over the years.
• Had to find ways to simplify data management
processes
• Other entities will also need to find ways to
simplify data management, may wish to
consider some of WECC’s methods
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
3
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
4
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
LID–Regions
CLZONE Climate Areas
NWC – Northwest coast
NWC` NWV – Northwest valley
NWI RMN NWI – Northwest inland
RMN – Rocky mountain
NWV
NCC – N. Calif. coast
NCV – N. Calif. valley
HID – High desert
SCC – S. Calif. coast
NCC
SCV – S. Calif. valley
HID
DSW – Desert southwest
NCV
SCC SCV
DSW
6
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
7
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
8
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
9
25
Static P Current
20 Static P
Resistive
Power
Electronic
15 Motor D
Motor C
Motor B
10
Motor A
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
10
0.30
0.25 Motor A
Motor B
0.20 Motor C
Motor D
Power Electronic
0.15
Static P Resistive
Static P Current
0.10
0.05
0.00
0 4 8 12 16 20 24
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
11
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
12
Spreadsheet Tool
5 5
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
14
Implementation Process
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
16
Implementation Process
• Phased implementation
– Currently disabled AC stalling – Phase 1
– Work continues to better understand AC stalling
– Hope to continue improving the model – Phase 2
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
17
W E S T E R N E L E C T R I C I T Y C O O R D I N A T I N G C O U N C I L
Is Load Loss Real ?
Presented by
Dmitry Kosterev, BPA
1
Hassayampa Event
• July 28, 2003 at 18:54
• 3-phase fault at Hassayampa 500-kV
substation west of Phoenix, AZ
• 2,685 MW of generation tripped following a
fault
3
40
30
20
10
Voltage (kV)
-10
-20
-30
-40
12 13 14 15 16 17 18 19 20 21 22
Time (cycles)
4
Pinnacle Peak frequency (pu)
60.05
60
59.95
59.9
59.85
59.8
59.75
50 70 90 110 130 150
60
59.95
59.9
59.85
59.8
59.75
50 55 60 65 70
7
Mid-Valley Event
• July 28, 2009 at 21:18
• Mid Valley 138-kV substation, Salt Lake City, UT
• Capacitor bank failure
• Fault initiated as a four cycle single phase to
ground fault that evolved into a three phase fault
for an additional six cycles. The fault was cleared
by action of the capacitor bank’s protective
relays. Total clearing time was about ten cycles.
9
15000
10000
5000
-5000
-10000
-15000
0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7
10
Summer Lake frequency (Hz)
60.1
60.08
60.06
60.04
60.02
60
59.98
59.96
-10 0 10 20 30 40 50 60
11
• Generation loss was about 190 MW
12
Modeling
24 V
Control
Commercial / residential loads:
– Motors B and C (fans and pumps)
• 20% trip at 60% voltage, reclose at 75%
• 30% trip at 50% voltage, reclose at 65%
– Motor A (compressors)
• 20% trip at 70% voltage and lock out
• 70% trip at 50% voltage, reclose at 70%
– Electronics
• Ramp down linearly as voltage declines from 70% to 50%
• 20% trip and remain off-line, 80% restart
15
Thank You
Motor Control and Protection
Control
Board
remains in
control Manufacturer Solid-state Control Board drops out
Over voltage, Manufacture
3-ph contactor below 65% V but will automatically restart. 300-
Phase r Solid-state
Compressor drops out at 500sec First Chiller 600sec interstaging delay for
Imbalance, Controller
Motors 50% V and each additional chiller if applicable
over current tied into EMS
re-energizes
Chillers after 1 to 8
cycles after
event
Over voltage, EMS & VFD operate through event then drops out
3-ph EMS & VFD
Phase EMS with 2 seconds after event below 65% V but will
Pump remains in
Imbalance, VFD automatically restart. Pump soft starts within 90
Motors control
over current seconds
20
Equipment Motors Protection Controls 5 cycle 10 cycle 2 second
cycle
Control
Board
remains in Manufacturer Solid-state Control Board drops out
Manufacture control below 65% V but will automatically restart.
1-ph r Solid-state contactor 120sec First Boiler 240sec interstaging delay for
Fuse &
Induced with drops out at each additional boiler if applicable
Thermal
Draft Motor contactor, 50% V and
EMS re-energizes Possible thermal trip if the voltage variance is
after 1 to 8 long enough and above 50% for 2-3 seconds
Boilers cycles after
event
Over voltage,
Phase EMS & VFD EMS & VFD operate through event then drops out
3-ph Imbalance, EMS with remains in 2 seconds after event below 65% V but will
Motors over current, VFD control automatically restart. Pump starts within 90
& current seconds
limiting
Over voltage, EMS & VFD operate through event then drops out
Phase 2 seconds after event below 65% V but will
EMS & VFD
Cooling 3-ph Imbalance, EMS with automatically restart. First fan starts within 5
remains in
Towers Fan Motor over current, VFD seconds 2nd fan if applicable re-starts at 30sec
control
& current
limiting
Conclusions
• We have developed a set of load response tables for a
range of commercial building types.
• We provide both the “drop out” and recovery
characteristic for typical motor loads, for the voltage
dips and times of interest, including both protection
and control components.
• This is a “next step” in better understanding of
commercial building load response to voltage
transients.
• http://www.pnnl.gov/main/publications/external/tech
nical_reports/PNNL-24468.pdf
• fidvr.lbl.gov
Additional Info Slides
VFDs
• Variable Frequency Drives are typically programmed to ride
through short duration voltage sags by current limiting the
motor. In cases where only one phase is sagging, and the
motor is being operated at partial load, the motor can run
for several seconds or more, depending on motor load.
• For 60% voltage and 5, 10 and 20 cycles, the VFD should be
able to ride through by current limiting the motor.
Depending on motor load, the VFD typically cannot ride
through a 2 second or 3 minute loss of voltage unless it is
equipped with energy storage.
• In testing, VFDs were noted to ride through sags of up to 2
seconds, or more, in duration, then trip after voltage
recovery.
Additional Info - Chiller Motors
2 RELIABILITY | ACCOUNTABILITY
Our System Load
Resistive Cooking
Resistive Heating
Power
Electronics
Distributed
Generation
Incandescent Lighting
Data Centers
AC and Heat
Pumps
Electric Vehicles
3 Share of total system load RELIABILITY | ACCOUNTABILITY
PTI PSS®E & GE PSLF Load Models
4 RELIABILITY | ACCOUNTABILITY
The CMLD (CMPLDW) Model
DG
GE PSLF
Siemens PTI PSS®E
Power World
PowerTech TSAT
5 RELIABILITY | ACCOUNTABILITY
CMPLDW/CMLD Debunked
6 RELIABILITY | ACCOUNTABILITY
The Distribution Equivalent Circuit
7 RELIABILITY | ACCOUNTABILITY
Motor A – Small Commercial
9 RELIABILITY | ACCOUNTABILITY
Motor A Model Data
10 RELIABILITY | ACCOUNTABILITY
Motor B
13 RELIABILITY | ACCOUNTABILITY
Motor D – Residential Air Conditioner
14 RELIABILITY | ACCOUNTABILITY
Motor D
15 RELIABILITY | ACCOUNTABILITY
Motor D – Performance
• Compressor Load
Torque is very cyclical
• Very possible that
motor stalls on next
compression cycle
• Compressor Motor
Inertia is very low
H = 0.03 – 0.05 sec
• Physically small
16 RELIABILITY | ACCOUNTABILITY
Motor D Model Representation
17 RELIABILITY | ACCOUNTABILITY
Motor D Performance Model
• Motors stall when voltage drops below Vstall for duration Tstall
• Fraction Frst of aggregate motor can restart when voltage
exceeds Vrst for duration Trst
5 5
3 3
2 2
RUN
0 0
0 0.2 0.4 0.6 0.8 1 1.2 0 0.2 0.4 0.6 0.8 1 1.2
Voltage (per unit) Voltage (per unit)
18 RELIABILITY | ACCOUNTABILITY
Thermal Relay Model
KTH
θ
1
IC2 *RSTALL 1 θ’ > 0
KTH
τTHs + 1 0 θ
θTRIP(1) θTRIP(2)
θ – compressor temperature
KTH – fraction of motors that remain connected
23 RELIABILITY | ACCOUNTABILITY
History of Load Modeling (in WECC)
24 RELIABILITY | ACCOUNTABILITY
History of Load Modeling (in WECC)
Model was used for 10+ years to plan and operate the Western
Interconnection
…Many utilities are choosing to use the CLOD model, which is similar to
this approach from 2001…!
25 RELIABILITY | ACCOUNTABILITY
SCE’s Observations and Modeling
• These models were used for special studies of local areas, but
beginning to get traction
27 RELIABILITY | ACCOUNTABILITY
WECC Load Modeling Task Force
29 PSS®E Model Library, Version 32.0.5, Revised October 2010. RELIABILITY | ACCOUNTABILITY
PTI PSS®E Load Models
30 PSS®E Model Library, Version 32.0.5, Revised October 2010. RELIABILITY | ACCOUNTABILITY
PTI PSS®E Load Models
• Aggregate Load
alwscc (b,w,z) – Load Voltage/Frequency Dependence Model
Secld1(2,3) – Secondary Load Model with Reset of Tap Ratio
• Induction Motor Load
apfl (spfl) – Pump/Fan Driven Induction (Synchronous) Motor Load Model
motorw/x – Single or Double Cage Induction Motor Model
• Single-phase Air Conditioner Load
Ld1pac – Performance-based Model of 1-φ Air Conditioner Load
motorc – Phasor Model of 1-φ Air Conditioner Load
• Other Loads
Ldelec (rect) – Electronic (Rectifier) Load Model
32 RELIABILITY | ACCOUNTABILITY
Substation & Feeder Parameters
33 RELIABILITY | ACCOUNTABILITY
T:D Transformer Parameters
34 RELIABILITY | ACCOUNTABILITY
Load Composition Parameters
35 RELIABILITY | ACCOUNTABILITY
Power Electronic Load Parameters
36 RELIABILITY | ACCOUNTABILITY
Static Real Load Parameters
37 RELIABILITY | ACCOUNTABILITY
Static Reactive Load Parameters
38 RELIABILITY | ACCOUNTABILITY
Motor Type Definition Parameters
39 RELIABILITY | ACCOUNTABILITY
Motor A Model Parameters
LpA 0.12
LppA 0.104
TpoA 0.095
TppoA 0.0021
HA 0.1 Majority of these motors are small – low inertia
etrqA 0* 𝑇𝑇𝑚𝑚𝑚𝑚𝑚𝑚𝑚 = 𝑇𝑇𝑚𝑚𝑚𝑚𝑚𝑚𝑚,0 ∗ 𝜔𝜔 𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡 - Constant Torque
*3φ motors driving constant torque loads (commercial air conditioner
compressors and refrigeration)
40 RELIABILITY | ACCOUNTABILITY
Motor A Protection Parameters
41 RELIABILITY | ACCOUNTABILITY
Motor B Model Parameters
LpB 0.19
LppB 0.14
TpoB 0.2
TppoB 0.0026
HB 0.5 Large inertia commercial/industrial fan motor loads
etrqB 2* 𝑇𝑇𝑚𝑚𝑚𝑚𝑚𝑚𝑚 = 𝑇𝑇𝑚𝑚𝑚𝑚𝑚𝑚𝑚,0 ∗ 𝜔𝜔 𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡 - Torque ∝ Speed-Squared
*3φ motors driving load proportional to speed-squared relationship with
high inertia (large fans)
42 RELIABILITY | ACCOUNTABILITY
Motor B Protection Parameters
43 RELIABILITY | ACCOUNTABILITY
Motor C Model Parameters
LpC 0.19
LppC 0.14
TpoC 0.2
TppoC 0.0026
HC 0.1 Large inertia commercial/industrial pump motor loads
etrqC 2* 𝑇𝑇𝑚𝑚𝑚𝑚𝑚𝑚𝑚 = 𝑇𝑇𝑚𝑚𝑚𝑚𝑚𝑚𝑚,0 ∗ 𝜔𝜔 𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡 - Torque ∝ Speed-Squared
*3φ motors driving load proportional to speed-squared relationship with
low inertia (pump loads)
44 RELIABILITY | ACCOUNTABILITY
Motor C Protection Parameters
45 RELIABILITY | ACCOUNTABILITY
Motor D Model Parameters
47 RELIABILITY | ACCOUNTABILITY
Motor D Protection Parameters
48 RELIABILITY | ACCOUNTABILITY
Motor D – Sensitivity to Ambient Temp
3.4 0.64
3.2 0.62
3 0.6
2.8 0.58
2.6 0.56
80 85 90 95 100 105 110 115
Ambient Temperature (F)
49 RELIABILITY | ACCOUNTABILITY
Current R&D Efforts
• Point-on-wave sensitivity
• Voltage sag rate-of-change sensitivity
Distribution recordings show sag is not
instantaneous
At least 1 cycle for voltage to sag – motor back-
feed
Vstall numbers lower than previously thought
50 RELIABILITY | ACCOUNTABILITY
Distribution FIDVR Monitoring
Richard Bravo
FIDVR Events
• PMUs has been recording FIDVR events for many years
• Limited information on how FIDVR events evolve in
distribution system
• Distribution FIDVR
events details needed
to assess:
– Spreading behavior
– Voltage levels at T&D
– Time of events
– Real and reactive
power demands
1
PQ Monitors on Residential Xmers
• Installed in pad-mount residential xmers secondary side: 240V
• Record residential loads aggregated behavior
• Record: V & I
– Line to
ground
voltage
– Line current
(aggregated)
2
PQ Monitors Installations
• Installed in Valley system dist. circuits (1,500 MW peak load)
• PQ threshold settings:
– UV triggers at
80%
– OV triggers at
110%
– Capture event
• RMS
• sinusoidal
waveforms
3
Event #1 (RMS)
• Multiple lightning strikes caused multiple distribution faults
recorded by the PQ devices, but not by transmission PMU
– P & Q increased
during FIDVR
• P=2.6 p.u. at V=90%
• Q=7 p.u. at V=90%
– FIDVR lasted 9 sec
– TOPs open
disconnecting loads
after seven (7)
second mark
• FIDVR recorded
only in distribution
system
4
Event #1 (sinusoidal)
• Fault initiated at ~70 degrees of voltage waveform
• Fault must have been in adjacent circuit
• Fault cleared fast but not fast enough to prevent A/C stalling
• Stalling prevented
voltage from
recovering
– Current waveform
(red) increases
significantly 200A
700A
– Voltage hold at 80%
– Current lagging
behavior increases
significantly during
the event
5
Event #6
• Lightning causes FIDVR event recorded by BOTH distribution
PQ devices and transmission PMU
6
DER Proposed VRT
• DER penetration is increasing significantly and may become a
major generating part of the grid during certain times
• Standards are being revised to allow voltage ride through
VOLTAGE MUST
RIDE-THROUGH
OPERATION DISCONNECT
(p.u.) (seconds)
(seconds)
>1.2 none Disconnect 0.16
Momentary
1.1 ~ 1.2 12 13
Cessation
0.88 ~ 1.10 Continuous Operation
Mandatory
0.70 ~ 0.88 20 21
Operation
Mandatory
0.5 ~ 0.7 10 11
Operation
Momentary
0 ~ 0.5 1 sec 1.5
Cessation
7
Ride Through Grid Voltage Events
8
DER Can Provide Grid Support
9
Conclusion
• No linear relationship between T&D voltages during FIDVR
• Faults at any point in the waveform can provoke FIDVR if
there is large induction motor load
• Stalling happens very quick within 2 cycles
• DER should ride through voltage events
• DER should supply VARs to support the voltage during
voltage events
• Voltage support typically less than 30 seconds so
minimum impact to generation revenew
10
Distribution Data for
FIDVR & Load Modeling
Kyle Thomas
ET Operations Engineering
Dominion Virginia Power
– NERC TPL-001
• Addition of dynamic load model requirement in planning studies
expected/planned
– Initial Simulation Observations
• Simulations using composite load model with best guess load composition
parameters show widespread FIDVR on Transmission system
• Actual Transmission level monitoring shows little to no widespread FIDVR
– Goal:
• Improve distribution level monitoring capability
• Use captured data to understand phenomena & model parameters
• Use improved model parameters to perform better informed studies
2
Distribution Data Gap – Capture Duration
SCADA
3
Distribution Data Gap – Monitoring Resolution
4
Initial Field Installations
2012-2014
5
Initial Field Installations
2012-2014
6
Initial Field Installations
2012-2014
7
Distribution PQ Meter Data
2013 and on
8
Example 1 – July 2006 4pm
Over-voltage ~ 1.2
pu
Rated Voltage
12.5kVLL
[kV L-N]
Voltage
10 s
Time [s]
9
Example 1 – July 2006 4pm
Line Current
[kA]
10 s
Time [s]
10
Example 2 – August 2006 6pm
Over-voltage ~ 1.1 pu
Rated Voltage
34.5kVLL
[kV L-N]
Voltage
10 s
Time [s]
11
Example 2 – August 2006 6pm
Line Current
[kA]
10 s
Time [s]
12
PQ Meter Settings
13
Latest Field Installations
Summer 2015
14
Latest Field Installations
Summer 2015
15
Going Forward
Standardization = Proliferation
16
Questions?
FIDVR
Voltage Dip Recordings
250 250
200 200
150 150
100 100
50 50
0 0
-0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 -1 0 1 2 3 4 5 6 7 8 9
Non-FIDVR events
300 300
280
280
260
260
240
220
240
Vl-n/V
Vl-n/V
200
220
180
160
200
140
180
120
160 100
-4 -2 0 2 4 6 8 10 12 14 16 -4 -2 0 2 4 6 8 10 12 14 16
Time/sec Time/sec
300 300
280
250
260
240
200
220
Vl-n/V
Vl-n/V 200
150
180
160
100
140
50 120
-4 -2 0 2 4 6 8 10 12 14 16 -4 -2 0 2 4 6 8 10 12 14 16
Time/sec Time/sec
Complicated events
350 290
280
300
270
250
260
250
200
Vl-n/V
Vl-n/V
240
150
230
220
100
210
50
200
0 190
-4 -2 0 2 4 6 8 10 12 14 16 -4 -2 0 2 4 6 8 10 12 14 16
Time/sec Time/sec
300 300
280
250
260
240 200
220
Vl-n/V
Vl-n/V
150
200
180 100
160
50
140
120 0
-4 -2 0 2 4 6 8 10 12 14 16 -4 -2 0 2 4 6 8 10 12 14 16
Time/sec Time/sec
Events of interest
300
280
260
240
220
Vl-n/V
200
180
160
140
120
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
Event 291
Dip Duration 0.117 second (7 cycles)
Initial voltage a,b,c 283.7 283.3 277.8
Final voltage a,b,c 287.4 281.1 277.7
Site10-2011-08-24 (T 19-07-02-574)
300
250
200
Volts L-N
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time, sec
Event 265
Dip Duration 0.067 second (4 cycles)
Initial voltage a,b,c 284.5 283.7 275.4
Final voltage a,b,c 296.4 280.3 277.7
Site10-2011-08-26 (T 17-16-46-130)
300
250
200
Volts L-N
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time, sec
Event 55
Dip Duration 0.4 second (24 cycles)
Initial voltage a,b,c 278.9 282.1 285.2
Final voltage a,b,c 289.2 296.9 298.5
Sites6and7-Site6-2013-08-23 (T 12-04-00-677)
300
250
200
Volts L-N
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time, sec
Event 243
Dip Duration 0.3 second (18 cycles)
Initial voltage a,b,c 279.4 284.5 286.5
Final voltage a,b,c 283.1 284.3 287.6
Site 6-2011-09-03 (T 11-30-57-915)
300
250
200
Volts L-N
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time, sec
Events 126 and 139
Dip Duration 0.4 second (24 cycles)
Init a,b,c 281.1 285.8 288.0 Init a,b,c 280.5 282.3 285.5
Final a,b,c 288.7 288.7 288.7 Final a,b,c 285.2 286.4 286.9
250 250
200 200
Volts L-N
Volts L-N
150 150
100 100
50 50
0 0
-4 -2 0 2 4 6 8 10 12 14 16 -4 -2 0 2 4 6 8 10 12 14 16
Time, sec Time, sec
Thanks to Centerpoint Energy
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
350
300
250
200
Vl-n/V
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16
Time/sec
Composite Load Model Sensitivity Study
2
Composite Load Model in WECC
Load Bus
M Motor C
M Motor D
UVLS
UFLS Electronic
Static
Study Work
4
Key Parameters (Motor D)
Vstall – Stall voltage, p.u. Vrst – Voltage at which restart can occur, p.u.
Tstall – Stall time delay, sec. Trst – Restart time delay
FmA – Motor A fraction of load P Vc1off – Contactor voltage at which tripping
starts, p.u.
FmB – Motor B fraction of load P
Vc2off – Contactor voltage at which tripping is
FmC – Motor C fraction of load P complete, p.u.
FmD – Motor D fraction of load P Vc1on – Contactor voltage at which
Fel – Electronic load fraction of P reconnection starts, p.u.
Vtr1 – First under voltage trip level, p.u. Vc2on – Contactor voltage at which
reconnection is complete, p.u.
Ttr1 – First under voltage trip delay time, sec.
Tth – Motor D thermal time constant, sec.
Fuvr – Fraction of load with under voltage
relay protection Th1t – Motor D thermal protection trip start
level, p.u. temperature
Frst – Fraction of load that can restart after
stalling Th2t – Motor D thermal protection trip
completion level, p.u. temperature
5
Key Parameters (Motors A, B, C)
6
Parameter Value Ranges
Phase 2
Ref. Base Value
Description of Parameter Phase 1 Minimum Maximum
No. as given in
Value Value
dyd/dyr
1 Vstall, Stall voltage, p.u. 0.5 0.5 0.3 0.8
2 Tstall, Stall time delay, sec. 9999 0.033 0.01667 0.25
3 Vc1off, Contactor voltage at which tripping starts, p.u. 0.5 0.5 0.3 0.7
4 Vc2off, Contactor voltage at which tripping is complete, p.u. 0.4 0.4 0.2 0.6
5 Vc1on, Contactor voltage at which reconnection is complete, p.u. 0.6 0.6 0.4 0.8
6 Vc2on, Contactor voltage at which reconnection starts, p.u. 0.5 0.5 0.3 0.7
7 Tth, Motor D thermal time constant, sec. 15 15 5 25
8 Th1t, Motor D thermal protection trip start level, p.u. temperature 0.7 0.7 0.4 0.9
9 Th2t, Motor D thermal protection trip completion level, p.u. temperature 1.2 1.2 1 3
10 FmA, Motor A fraction of load P 0.167 0.167 -20% +20%
11 FmB, Motor B fraction of load P 0.135 0.135 -20% +20%
12 FmC, Motor C fraction of load P 0.061 0.061 -20% +20%
13 FmD, Motor D fraction of load P 0.113 0.113 -20% +20%
14 Fel, Electronic load fraction of P 0.173 0.173 -20% +20%
15 Vtr1, First under voltage trip level, p.u. 0.6 0.6 0.4 0.8
16 Ttr1, First under voltage trip delay time, sec. 0.02 0.02 0.01667 0.25
17 Fuvr, Fraction of load with under voltage relay protection 0.1 0.1 0 0.5
18 Frst, Fraction of load that can restart after stalling 0.2 0.2 0 1
19 Vrst, Voltage at which restart can occur, p.u. 0.95 0.95 0.5 1
20 Trst, Restart time delay 0.3 0.3 0.1 1
Notes:
1. Contactor settings (Vc1off, Vc2off, Vc1on, Vc2on) are changed simultaneously.
2. Motor fraction (FmA, FmB, FmC, FmD) base values are examples. Load fractions vary from load to load in the base case.
7
FmD (Voltages)
8
FmA (Voltages)
9
Tstall (Voltages)
10
Tth (Voltages)
11
Sensitivity Metrics
12
Study Criteria (Thresholds)
• Transient voltage dips will be monitored and recorded for dips that
exceed:
25% at load buses and 30% at non-load buses
20% for more than 20 cycles at load buses
• Frequency oscillations will be monitored and recorded for dips below
59.6 Hz for 6 cycles or more at load buses
• Post-transient voltage deviations exceeding 5% at any bus
• Voltage recovery to 70% in 1 second, 80% in 3 seconds, 90% in 5 seconds
• Power, angle, or voltage magnitude oscillations will be monitored and
recorded for 5% damping from the first swing peak to the 3rd swing peak
• Voltage overshoot will be monitored
Voltage greater than 1.1 p.u, or greater than 1.05 p.u. for 5 seconds or longer.
• Any non-consequential load loss (total load loss to be calculated)
• Any generator that loses synchronism
• Damping ratio sensitivities (stressed cases)
13
Sensitivity Results – Utility A (1)
Voltage overshoot
Voltage < 70% in 1 second Voltage < 80% in 3 seconds Voltage < 90% in 5 seconds
Ref. over 1.1 p.u.
Variable Setting
No. Delta Delta Delta Delta
# of Buses # of Buses # of Buses # of Buses
(% change) (% change) (% change) (% change)
1 Phase 2 - 78 - 1011 - 4506 - 3194 -
2 -20% 22 -72% 946 -6% 4450 -1% 2503 -22%
Fel
3 +20% 164 110% 1035 2% 4715 5% 3634 14%
4 -20% 19 -76% 1255 24% 5150 14% 2153 -33%
FmA
5 +20% 211 171% 882 -13% 4151 -8% 4582 43%
6 -20% 20 -74% 1043 3% 4631 3% 2523 -21%
FmB
7 +20% 151 94% 1049 4% 4392 -3% 3480 9%
8 -20% 22 -72% 1076 6% 4552 1% 2811 -12%
FmC
9 +20% 101 29% 1005 -1% 4558 1% 3466 9%
10 -20% 19 -76% 415 -59% 2928 -35% 2015 -37%
FmD
11 +20% 219 181% 1623 61% 5504 22% 3829 20%
12 Frst 0 78 0% 1011 0% 4518 0% 3302 3%
13 (0.2) 1 78 0% 1011 0% 4518 0% 2930 -8%
14 Fuvr 0 78 0% 1472 46% 5637 25% 2571 -20%
15 (0.1) 0.5 78 0% 27 -97% 66 -99% 6428 101%
16 Th1t 0.4 78 0% 1011 0% 4432 -2% 3199 0%
17 (0.7) 0.9 78 0% 1011 0% 4506 0% 3142 -2%
18 Th2t 1 78 0% 1011 0% 4505 0% 5196 63%
19 (1.2) 3 78 0% 1011 0% 4506 0% 1406 -56%
Notes:
1. Phase 2 base case values are shown in parentheses under each parameter.
14
Sensitivity Results – Utility A (2)
Voltage overshoot
Voltage < 70% in 1 second Voltage < 80% in 3 seconds Voltage < 90% in 5 seconds
Ref. over 1.1 p.u.
Variable Setting
No. Delta Delta Delta Delta
# of Buses # of Buses # of Buses # of Buses
(% change) (% change) (% change) (% change)
20 Trst 0.1 78 0% 1011 0% 4518 0% 3176 -1%
21 (0.3) 1 78 0% 1011 0% 4518 0% 3265 2%
22 Tstall 0.25 78 0% 27 -97% 62 -99% 8600 169%
23 (0.033) 0.01667 210 169% 5404 435% 13919 209% 8121 154%
24 Tth 5 78 0% 1011 0% 4148 -8% 5948 86%
25 (15) 25 78 0% 1011 0% 4506 0% 2668 -16%
26 Ttr1 0.25 78 0% 1452 44% 5578 24% 2674 -16%
27 (0.02) 0.01667 78 0% 1020 1% 4501 0% 3292 3%
28 Vc1off 0.7 18 -77% 133 -87% 3392 -25% 3599 13%
29 (0.5) 0.3 148 90% 2693 166% 7498 66% 3815 19%
30 Vrst 1 78 0% 1011 0% 4518 0% 3276 3%
31 (0.95) 0.5 78 0% 340 -66% 3477 -23% 3049 -5%
32 Vstall 0.3 78 0% 27 -97% 171 -96% 7688 141%
33 (0.5) 0.8 4822 6082% 32667 3131% 49106 990% 25291 692%
34 Vtr1 0.4 78 0% 1385 37% 5622 25% 2588 -19%
35 (0.6) 0.8 20 -74% 999 -1% 4763 6% 3179 0%
Notes:
1. Phase 2 base case values are shown in parentheses under each parameter.
2. Contactor settings (Vc1off, Vc2off, Vc1on, Vc2on) are changed simultaneously. (Ref. No. 28 and 29)
15
Sensitivity Results – Utility B (1)
Voltage overshoot
Voltage < 70% in 1 second Voltage < 80% in 3 seconds Voltage < 90% in 5 seconds
Ref. over 1.1 p.u.
Variable Setting
No. Delta Delta Delta Delta
# of Buses # of Buses # of Buses # of Buses
(% change) (% change) (% change) (% change)
1 Phase 2 - 10770 - 17866 - 23250 - 6717 -
2 -20% 10327 -4% 16271 -9% 21092 -9% 6160 -8%
Fel
3 +20% 10938 2% 19003 6% 24204 4% 7974 19%
4 -20% 11351 5% 16923 -5% 21160 -9% 4150 -38%
FmA
5 +20% 9818 -9% 18353 3% 24147 4% 9589 43%
6 -20% 9493 -12% 14016 -22% 17921 -23% 5018 -25%
FmB
7 +20% 11067 3% 19909 11% 25915 11% 8243 23%
8 -20% 10485 -3% 16572 -7% 21099 -9% 5838 -13%
FmC
9 +20% 10845 1% 18707 5% 23906 3% 7813 16%
10 -20% 8219 -24% 12982 -27% 17528 -25% 14550 117%
FmD
11 +20% 11259 5% 17608 -1% 23858 3% 1741 -74%
12 Frst 0 10770 0% 17910 0% 23220 0% 6478 -4%
13 (0.2) 1 10770 0% 17870 0% 23182 0% 7928 18%
14 Fuvr 0 12235 14% 22019 23% 26668 15% 2533 -62%
15 (0.1) 0.5 2899 -73% 3916 -78% 7354 -68% 15373 129%
16 Th1t 0.4 10770 0% 17884 0% 23243 0% 21466 220%
17 (0.7) 0.9 10770 0% 17851 0% 23183 0% 1062 -84%
18 Th2t 1 10770 0% 17880 0% 23254 0% 31083 363%
19 (1.2) 3 10770 0% 17862 0% 23252 0% 185 -97%
Notes:
1. Phase 2 base case values are shown in parentheses under each parameter.
16
Sensitivity Results – Utility B (2)
Voltage overshoot
Voltage < 70% in 1 second Voltage < 80% in 3 seconds Voltage < 90% in 5 seconds
Ref. over 1.1 p.u.
Variable Setting
No. Delta Delta Delta Delta
# of Buses # of Buses # of Buses # of Buses
(% change) (% change) (% change) (% change)
20 Trst 0.1 10770 0% 17851 0% 23234 0% 6961 4%
21 (0.3) 1 10550 -2% 17634 -1% 23148 0% 6818 2%
22 Tstall 0.25 3439 -68% 24 -100% 30 -100% 9206 37%
23 (0.033) 0.01667 14448 34% 25039 40% 30774 32% 12215 82%
24 Tth 5 10769 0% 17852 0% 23181 0% 29807 344%
25 (15) 25 10770 0% 17829 0% 23246 0% 185 -97%
26 Ttr1 0.25 12235 14% 21227 19% 26173 13% 9984 49%
27 (0.02) 0.01667 10896 1% 18238 2% 23555 1% 7056 5%
28 Vc1off 0.7 2791 -74% 3248 -82% 9387 -60% 13917 107%
29 (0.5) 0.3 17262 60% 29810 67% 33771 45% 1764 -74%
30 Vrst 1 10770 0% 17870 0% 23253 0% 6565 -2%
31 (0.95) 0.5 10770 0% 17939 0% 23667 2% 14915 122%
32 Vstall 0.3 5474 -49% 2532 -86% 3466 -85% 2915 -57%
33 (0.5) 0.8 19832 84% 33328 87% 39705 71% 38333 471%
34 Vtr1 0.4 12042 12% 21813 22% 26404 14% 3602 -46%
35 (0.6) 0.8 10636 -1% 14462 -19% 19165 -18% 4893 -27%
Notes:
1. Phase 2 base case values are shown in parentheses under each parameter.
2. Contactor settings (Vc1off, Vc2off, Vc1on, Vc2on) are changed simultaneously. (Ref. No. 28 and 29)
17
Sensitivity Results – Utility C (1)
Voltage overshoot
Voltage < 70% in 1 second Voltage < 80% in 3 seconds Voltage < 90% in 5 seconds
Ref. over 1.1 p.u.
Variable Setting
No. Delta Delta Delta Delta
# of Buses # of Buses # of Buses # of Buses
(% change) (% change) (% change) (% change)
1 Phase 2 - 2127 - 2025 - 3092 - 2680 -
2 -20% 2097 -1% 2087 3% 3050 -1% 2370 -12%
Fel
3 +20% 2132 0% 2170 7% 3289 6% 2875 7%
4 -20% 2322 9% 2256 11% 3554 15% 2657 -1%
FmA
5 +20% 1925 -9% 1959 -3% 3021 -2% 2946 10%
6 -20% 2066 -3% 1990 -2% 2993 -3% 2373 -11%
FmB
7 +20% 2150 1% 2246 11% 3362 9% 2934 9%
8 -20% 2106 -1% 2016 0% 3059 -1% 2493 -7%
FmC
9 +20% 2096 -1% 2100 4% 3232 5% 2793 4%
10 -20% 1960 -8% 1524 -25% 2616 -15% 1846 -31%
FmD
11 +20% 2137 0% 2297 13% 3311 7% 2485 -7%
12 Frst 0 2127 0% 2020 0% 3104 0% 2733 2%
13 (0.2) 1 2127 0% 2019 0% 3088 0% 2670 0%
14 Fuvr 0 2433 14% 2738 35% 4075 32% 2655 -1%
15 (0.1) 0.5 1332 -37% 359 -82% 920 -70% 4248 59%
16 Th1t 0.4 2128 0% 2018 0% 3058 -1% 2778 4%
17 (0.7) 0.9 2128 0% 2019 0% 3094 0% 1825 -32%
18 Th2t 1 2127 0% 2019 0% 3096 0% 5426 102%
19 (1.2) 3 2128 0% 2019 0% 3094 0% 174 -94%
Notes:
1. Phase 2 base case values are shown in parentheses under each parameter.
18
Sensitivity Results – Utility C (2)
Voltage overshoot
Voltage < 70% in 1 second Voltage < 80% in 3 seconds Voltage < 90% in 5 seconds
Ref. over 1.1 p.u.
Variable Setting
No. Delta Delta Delta Delta
# of Buses # of Buses # of Buses # of Buses
(% change) (% change) (% change) (% change)
20 Trst 0.1 2127 0% 2019 0% 3096 0% 2624 -2%
21 (0.3) 1 2127 0% 2019 0% 3096 0% 2736 2%
22 Tstall 0.25 1264 -41% 106 -95% 59 -98% 5079 90%
23 (0.033) 0.01667 2882 35% 3104 53% 5100 65% 4877 82%
24 Tth 5 2127 0% 2010 -1% 2905 -6% 5361 100%
25 (15) 25 2127 0% 2019 0% 3097 0% 204 -92%
26 Ttr1 0.25 2432 14% 2676 32% 3985 29% 3308 23%
27 (0.02) 0.01667 2158 1% 2065 2% 3097 0% 2679 0%
28 Vc1off 0.7 1101 -48% 559 -72% 1291 -58% 2992 12%
29 (0.5) 0.3 3165 49% 3663 81% 5533 79% 3580 34%
30 Vrst 1 2127 0% 2019 0% 3096 0% 2722 2%
31 (0.95) 0.5 2125 0% 1898 -6% 3042 -2% 2156 -20%
32 Vstall 0.3 1748 -18% 264 -87% 418 -86% 3578 34%
33 (0.5) 0.8 4271 101% 6946 243% 10086 226% 8070 201%
34 Vtr1 0.4 2415 14% 2726 35% 4126 33% 2804 5%
35 (0.6) 0.8 2369 11% 2140 6% 3500 13% 3152 18%
Notes:
1. Phase 2 base case values are shown in parentheses under each parameter.
2. Contactor settings (Vc1off, Vc2off, Vc1on, Vc2on) are changed simultaneously. (Ref. No. 28 and 29)
19
Overall Summary
• Observed similar results for all sets of data (SRP, SCE,
PG&E, PacifiCorp)
20
Other Observations
21
Next Steps
• Discuss results with each Transmission Provider
Summary of results and sensitivity tables
Detailed results for each contingency can be provided
Select contingencies for more detailed analysis
• Criteria for selecting contingencies to be documented
22
BPA Studies Using Composite Load
Model – Portland Metro Area
Presented by
Dmitry Kosterev, BPA
1
BPA Overview
• Bonneville Power
Administration (BPA) is a
federal Power Marketing
Agency in Pacific Northwest
• BPA markets power from 31
Federal dams and the
Columbia Generating Station
Nuclear Plant
• BPA operates more than
15,000 miles of
transmission, including
4,735 miles of 500-kV lines
2
Portland Area Study
The study has multiple objectives:
• Modeling
– One of the first large scale studies using phase 2 load models
(air-conditioner stalling is enabled)
• Reliability Assessment
• What types of faults and under what conditions can cause load-
induced voltage instability or delayed voltage recovery in Portland
Metro area?
• Should a wide-spread load-induced voltage instability or delayed
voltage recovery occur in Portland Metro area, what are the risks of it
cascading in other parts of the system?
• What solutions can be used to mitigate FIDVR phenomenon and limit
its propagation?
4
… but not very often
PDX Temperature
105
2005
2006
2007
2008
2009
100
2010
2011
2012
Temperature (F)
2013
2014
95 2015
90
85
0 50 100 150
Hours
5
Temperature
110 Sub-urban
residential
100
90
neighborhood
80
70
60
50
(newer
construction)
0 3 6 9 12 15 18 21 24
0 6
0 3 6 9 12 15 18 21 24
Temperature
110
Downtown
commercial
100
90
80
70
60
50
0 3 6 9 12 15 18 21 24
30
25
20
15
10
Commercial
5
0
loads show less
0 3 6 9 12 15 18 21 24
temperature
sensitivity
E-substation - Reactive Power
14
12
10
0 7
0 3 6 9 12 15 18 21 24
Temperature
110
100
90
Mixed Loads
80
70
60
50
0 3 6 9 12 15 18 21 24
25
20
15
10
0
0 3 6 9 12 15 18 21 24
10 Motor B
Motor A
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
0.25
Motor A
0.20 Motor B
Motor C
Motor D
0.15
Power Electronic
Static P Resistive
0.10 Static P Current
0.05
0.00 10
0 4 8 12 16 20 24
Modeling Assumptions (for HE 17:00)
Normal Summer
Load MA MB MC MD Electronic Static - R Static - I
Mixed 0.16 0.15 0.05 0.17 0.15 0.16 0.15
Residential 0.08 0.13 0.03 0.31 0.13 0.23 0.09
Commercial 0.21 0.14 0.03 0.12 0.19 0.12 0.19
1.15
1.1
3-phase faults
1.05 500-kV and 230-kV
normal clearing
1
0.95
0.9
0.85
0.8
0.75
0.7
0 5 10 15 20 25 30
1.15
1.1
1.05
0.95
0.9
0.85
0.8
0.75
0.7
0 5 10 15 20 25 30
12
California – Oregon Intertie
Malin 500-kV Bus Voltage
0.7
0.6
0 5 10 15 20 25 30
7000
6000
5000
4000
3000
2000
1000
0 5 10 15 20 25 30
13
Portland Area SVC
Keeler 230-kV Bus Voltage
0.7
0.6
0 5 10 15 20 25 30
300
200
100
-100
-200
-300
-400
-500
0 5 10 15 20 25 30
14
Portland Area Generation
River Road Stator Voltage River Road Field Current
3.5
1.1
3
1
2.5
0.9
2
0.8
1.5
0.7
0.6 1
0 5 10 15 20 25 30 0 5 10 15 20 25 30
300 300
250 200
200 100
150 0
100 -100
50 -200
0 -300
0 5 10 15 20 25 30 0 5 10 15 20 25 30
15
Fault Type Matters
3-phase fault, normal clearing 1-phase fault, normal clearing
500-kV Bus Voltage
500-kV Bus Voltage
1.15
1.15
1.1
1.1
1.05
1.05
1
1
0.95
0.95
0.9
0.9
0.85
0.85
0.8
0.8
0.75
0.75
0.7
0.7 0 5 10 15 20 25 30
0 5 10 15 20 25 30
1.15
1.15
1.1
1.1
1.05
1.05
1
1
0.95
0.95
0.9
0.9
0.85
0.85
0.8
0.8
0.75
0.75
0.7
0.7 0 5 10 15 20 25 30
0 5 10 15 20 25 30
1.15
1.15
1.1
1.1
1.05
1.05
1
1
0.95
0.95
0.9
0.9
0.85
0.85
0.8
0.8
0.75 0.75
0.7 0.7
0 5 10 15 20 25 30 0 5 10 15 20 25 30
1.15 1.15
1.1 1.1
1.05 1.05
1 1
0.95 0.95
0.9 0.9
0.85 0.85
0.8 0.8
0.75 0.75
0.7 0.7
0 5 10 15 20 25 30
0 5 10 15 20 25 30
19
Next Steps – Portland Area Studies
• Monitoring
• BPA has good PMU coverage of 500-kV grid and
main 230-kV substations in the area
• Continue expansion of synchronzied
measurements down to sub-transmission and
distribution levels (work with LSEs)
• Model Improvements
• Re-run the studies after the revised AC model is
implemented – results are expected to get better
• Continue risk-based scenario planning to ensure
robustness with respect to unexpected generator
control actions and trips during FIDVR
20
Thank You
21
National Grid Experience with PSS/E Composite
Load Model
1
CMLD Model
2
Baseline CMLD Parameters - Load Breakdown
Constant Constant
New England Region Electronics Motor A Motor B Motor C Motor D Current Impedance
Connecticut 18% 14% 12% 6% 25% 12% 13%
Massachusetts – East 16% 18% 12% 7% 23% 12% 13%
Massachusetts -
West/Central 14% 15% 13% 8% 25% 10% 14%
Maine 16% 15% 12% 9% 19% 12% 17%
New Hampshire 16% 16% 12% 8% 18% 13% 17%
Rhode Island 14% 15% 13% 7% 26% 11% 14%
Vermont 15% 17% 11% 10% 19% 12% 16%
New England 16% 16% 12% 7% 23% 12% 14%
Baseline CMLD Parameters – Motor A
9
Sensitivity Testing: Motor D (1-phase Air Conditioner)
Varied stall voltage (Vstall):
Baseline: Vstall = 0.40 pu
Sensitivity 1: Vstall = 0.35 pu
Sensitivity 2: Vstall = 0.30 pu
12
Results for lowering VC1off and VC2off for Motor D:
Vstall = 40%, VC1off = 0.40 pu, VC2off = 0.30 pu
13
Results for lowering VC1off and VC2off for Motor D:
Vstall = 40%, VC1off = 0.35 pu, VC2off = 0.25 pu
14
Results using Baseline assumptions for Motor D:
Vstall = 35%, VC1off = 0.45 pu, VC2off = 0.35 pu
15
Results using Baseline assumptions for Motor D:
Vstall = 35%, VC1off = 0.40 pu, VC2off = 0.30 pu
16
Results using Baseline assumptions for Motor D:
Vstall = 35%, VC1off = 0.35 pu, VC2off = 0.25 pu
17
Results using Baseline assumptions for Motor D:
Vstall = 30%, VC1off = 0.40 pu, VC2off = 0.30 pu
18
Results using Baseline assumptions for Motor D:
Vstall = 30%, VC1off = 0.35 pu, VC2off = 0.25 pu
19
Sensitivity to Motor A, B, & C
Under-voltage Contactor Dropout
Voltages
20
Sensitivity Testing: Motor A, B, C U/V Trip Parameters
Baseline assumptions (Kosterev) Motor A Motor B Motor C
Vtr1 - U/V Trip1 V (pu) 0.70 pu 0.50 pu 0.70 pu
Ftr1 - U/V Trip1 fraction 0.20 0.50 0.20
Vrc1 - U/V Trip1 reclose V (pu) no reclose 0.70 pu no reclose
Vtr2 - U/V Trip2 V (pu) 0.50 pu 0.50 pu 0.50 pu
Ftr2 - U/V Trip2 fraction 0.70 0.50 0.70
Vrc2 - U/V Reclose V (pu) 0.70 pu no reclose 0.70 pu
Sensitivity parameters
Vtr1 - U/V Trip1 V (pu) 0.70 pu 0.70 pu 0.70 pu
Ftr1 - U/V Trip1 fraction 0.50 0.50 0.50
Vrc1 - U/V Trip1 reclose V (pu) no reclose no reclose no reclose
Vtr2 - U/V Trip2 V (pu) 0.50 pu 0.50 pu 0.50 pu
Ftr2 - U/V Trip2 fraction 0.50 0.50 0.50
Vrc2 - U/V Reclose V (pu) 0.95 pu 0.95 pu 0.95 pu
Load loss using Baseline Assumptions
22
Load loss using with Motor A, B, C U/V Tripping
Sensitivity Parameters
23
Numerical Problems
With CMPL Model in PSSE
24
120 Hz oscillations found to certain 3ph fault
120 Hz Oscillation
6.25 Hz Oscillation
25
Changed Motor A parameter LPPA from 0.104 pu to 0.12 pu
27
Changed Network Solution Iterations to 200
29
Questions?
30
Experience with CMLD in a
Practical SOL Application
DOE-NERC
FIDVR & Dynamic Load Modeling Conference
Alexandria, VA
October 1, 2015
Robert J. O’Keefe
American Electric Power
Load Area Overview
120 miles to
interconnected
system
345 kV 345 kV
138 kV 138 kV
138 kV
G G 138-69 kV
100 miles
Network G
to another
load area
Single-Phase A/C Component
Sensitivity
Transition to
Transition to
Stall Mode
Stall Mode
P
P
Q
Q
EPRI CMLD Data
Steady-State P,Q vs V Characteristics
D-component
Transition
toTransition
Stall to P P
Stall Mode
Mode
C-Component
C-component
100% Stall/Trip
100% Stall/Trip
Q
Q
EPRI CMLD Data, D-Component = 0%
Steady-State P,Q vs V Characteristics
C-component
stalls but PP
doesC-component
not trip stalls
but does not trip
Q
Q
EPRI CMLD Data, D-Component = 100%
Tstall = .033
P P
V infinite
V infinite
bus
bus
QQ
VVload
load
bus
bus
EPRI CMLD Data, D-Component = 100%
Tstall = 999
PP
VVinfinite
infinite
bus
bus
QQ
V
V load
load
bus
bus
EPRI CMLD Data, D-Component = 100%
Vstall = 0.10
PSS/E Implementation Questions
P P
Q
Q
Other CMLD Advice
Oct 1, 2015
Jun Wen
Southern California Edison
jun.wen@sce.com
EDISON INTERNATIONAL® 1
SM
EDISON INTERNATIONAL® 2
SM
EDISON INTERNATIONAL® 3
SM
MOTORW
CMPLDW
4 cycle 3-phase fault @ Valley 500 kV
Trip Valley-Serrano 500 kV line
EDISON INTERNATIONAL® 4
SM
EDISON INTERNATIONAL® 5
SM
EDISON INTERNATIONAL® 6
SM
Fault location
Note: Total 80 load buses in SCE’s service territory are modeled with composite load models
EDISON INTERNATIONAL® 7
SM
Vtr1 Ttr1 Ftr1 Vrc1 Trc1 Vtr2 Ttr2 Ftr2 Vrc2 Trc2
MA 0.6 0.7 0.02 0.033 0.2 0.3 1 9999 9999 0.5 0.5 0.02 0.033 0.6 0.7 0.65 0.1 0.2
MB 0.6 0.7 0.02 0.033 0.25 0.35 0.9 0.05 9999 0.5 0.6 0.02 0.033 0.5 0.6 0.85 0.05 0.2
MC 0.6 0.7 0.02 0.033 0.25 0.35 0.9 0.05 9999 0.5 0.6 0.02 0.033 0.5 0.6 0.85 0.05 0.2
* Note: black – default, red - sensitivity
EDISON INTERNATIONAL® 8
SM
Voltage @ Chino 66 kV
Case 2: MA Ftr1=0.3
Case 6: MA, MB, MC Ftr1=0.3, 0.35, 0.35
EDISON INTERNATIONAL® 9
SM
EDISON INTERNATIONAL® 10
SM
EDISON INTERNATIONAL® 11
SM
EDISON INTERNATIONAL® 12
SM
1.0
0.8
kV (pu)
0.6
0.4
0.2
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Time (seconds)
EDISON INTERNATIONAL® 13
SM
1.0
0.8
kV (pu)
0.6
0.4
0.2
0.0
0.0 5.0 10.0 15.0 20.0 25.0 30.0
Time (seconds)
With default data set, the simulation showed a close match to PMU measurement
(initial voltage recovery, post voltage recovery, and load loss). The simulation shows
faster recovery than measurement (Tth can be modified to fit the curve).
EDISON INTERNATIONAL® 14
SM
EDISON INTERNATIONAL® 15
SM
Single-phase to ground
fault at time 0;
Blue – unsuccessful
reclosing at time 1 second
EDISON INTERNATIONAL® 16
SM
THANK YOU
EDISON INTERNATIONAL® 17
Load Modeling & FIDVR
FIDVR & Dynamic Load Modeling Workshop
September 30 – October 1, 2015
Robert W. Cummings
NERC Director of Reliability Initiatives and System Analysis
Introduction
3 RELIABILITY | ACCOUNTABILITY
Distributed Generation
5 RELIABILITY | ACCOUNTABILITY
Nature of the Issue
*http://www.nerc.com/files/glossary_of_terms.pdf
7 RELIABILITY | ACCOUNTABILITY
Voltage Sensitivities Highlighted
FROM… TO…
P=P0[p1V2+p2V+p3] • 3-phase Motors – Fans, Pumps, Compressors
Q=Q0[q1V2+q2V+q3] • 1-phase Induction Motors
• Power Electronic Load
• Static (Polynomial) Load
9 • Distribution Equivalent RELIABILITY | ACCOUNTABILITY
Transient Voltage Recovery
10 RELIABILITY | ACCOUNTABILITY
Goals for Reliability
11 RELIABILITY | ACCOUNTABILITY
Moving Forward – Next Steps
• Share best practices for dynamic load modeling and FIDVR events
• Share best practices for non-traditional resource modeling
• Collaborate with software vendors to further develop and
improve available dynamic models in software
• Continue engaging manufacturing community to raise awareness
of grid needs – promote grid-friendly devices
• Engage in IEEE equipment standards – awareness of aggregate
impact of multiple small devices
• Collect as much load data as possible (classification, end use ,
feeder information, etc.)
• Develop a process for creating load models – zonal or regional
load models are NOT sufficient
• Sensitivity, sensitivity, and more sensitivity studies
12 RELIABILITY | ACCOUNTABILITY
13 RELIABILITY | ACCOUNTABILITY
U.S. DOE-NERC Workshop on
Fault-Induced Delayed Voltage
Recovery (FIDVR) & Dynamic Load
Modeling
Summary
Page 6
10-7-10